Holdout version

Serine distribution in features of Nuclear_model

L18 motif ngrams in Membrane model

## [1] "Number of all features: 11391"
## [1] "Number of features contained within L18 motif: 79"
## [1] "Number of features contained within DPLG motif: 10"

L18 motif ngrams in Nuclear membrane model

## [1] "Number of all features: 4062"
## [1] "Number of features contained within L18 motif: 105"
## [1] "Number of features contained within DPLG motif: 13"

## L18 motif ngrams in N_E vs. N_TM model

## [1] "Number of all features: 649"
## [1] "Number of features contained within L18 motif: 11"
## [1] "Number of features contained within DPLG motif: 3"

Features with the highest Gini importance

## [[1]]

## 
## [[2]]

## 
## [[3]]

## 
## [[4]]

## 
## [[5]]

## 
## [[6]]

All features with Gini importance > 0

Partitioning version

Serine distribution in features of Nuclear_model

L18 motif ngrams in Membrane model

## [1] "Number of all features: 12035"
## [1] "Number of features contained within L18 motif: 74"
## [1] "Number of features contained within DPLG motif: 9"

L18 motif ngrams in Nuclear membrane model

## [1] "Number of all features: 4443"
## [1] "Number of features contained within L18 motif: 105"
## [1] "Number of features contained within DPLG motif: 13"

## L18 motif ngrams in N_E vs. N_TM model

## [1] "Number of all features: 848"
## [1] "Number of features contained within L18 motif: 13"
## [1] "Number of features contained within DPLG motif: 5"

Features with the highest Gini importance

## [[1]]

## 
## [[2]]

## 
## [[3]]

## 
## [[4]]

## 
## [[5]]

## 
## [[6]]

All features with Gini importance > 0